From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: richard.henderson@linaro.org
Subject: [PATCH] target/i386: fix operand size of unary SSE operations
Date: Mon, 9 Jan 2023 17:16:31 +0100 [thread overview]
Message-ID: <20230109161635.276783-1-pbonzini@redhat.com> (raw)
VRCPSS, VRSQRTSS and VCVTSx2Sx have a 32-bit or 64-bit memory operand,
which is represented in the decoding tables by X86_VEX_REPScalar. Add it
to the tables, and make validate_vex() handle the case of an instruction
that is in exception type 4 without the REP prefix and exception type 5
with it; this is the cas of VRCP and VRSQRT.
Reported-by: yongwoo <https://gitlab.com/yongwoo36>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1377
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
target/i386/tcg/decode-new.c.inc | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/target/i386/tcg/decode-new.c.inc b/target/i386/tcg/decode-new.c.inc
index 80c579164ff2..d5fd8d965c44 100644
--- a/target/i386/tcg/decode-new.c.inc
+++ b/target/i386/tcg/decode-new.c.inc
@@ -105,6 +105,7 @@
#define vex3 .vex_class = 3,
#define vex4 .vex_class = 4,
#define vex4_unal .vex_class = 4, .vex_special = X86_VEX_SSEUnaligned,
+#define vex4_rep5 .vex_class = 4, .vex_special = X86_VEX_REPScalar,
#define vex5 .vex_class = 5,
#define vex6 .vex_class = 6,
#define vex7 .vex_class = 7,
@@ -839,8 +840,8 @@ static const X86OpEntry opcodes_0F[256] = {
[0x50] = X86_OP_ENTRY3(MOVMSK, G,y, None,None, U,x, vex7 p_00_66),
[0x51] = X86_OP_GROUP3(sse_unary, V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
- [0x52] = X86_OP_GROUP3(sse_unary, V,x, H,x, W,x, vex5 p_00_f3),
- [0x53] = X86_OP_GROUP3(sse_unary, V,x, H,x, W,x, vex5 p_00_f3),
+ [0x52] = X86_OP_GROUP3(sse_unary, V,x, H,x, W,x, vex4_rep5 p_00_f3),
+ [0x53] = X86_OP_GROUP3(sse_unary, V,x, H,x, W,x, vex4_rep5 p_00_f3),
[0x54] = X86_OP_ENTRY3(PAND, V,x, H,x, W,x, vex4 p_00_66), /* vand */
[0x55] = X86_OP_ENTRY3(PANDN, V,x, H,x, W,x, vex4 p_00_66), /* vandn */
[0x56] = X86_OP_ENTRY3(POR, V,x, H,x, W,x, vex4 p_00_66), /* vor */
@@ -878,7 +879,7 @@ static const X86OpEntry opcodes_0F[256] = {
[0x58] = X86_OP_ENTRY3(VADD, V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
[0x59] = X86_OP_ENTRY3(VMUL, V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
- [0x5a] = X86_OP_GROUP3(sse_unary, V,x, H,x, W,x, vex3 p_00_66_f3_f2),
+ [0x5a] = X86_OP_GROUP3(sse_unary, V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
[0x5b] = X86_OP_GROUP0(0F5B),
[0x5c] = X86_OP_ENTRY3(VSUB, V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
[0x5d] = X86_OP_ENTRY3(VMIN, V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
@@ -1447,9 +1448,9 @@ static bool validate_vex(DisasContext *s, X86DecodedInsn *decode)
* Instructions which differ between 00/66 and F2/F3 in the
* exception classification and the size of the memory operand.
*/
- assert(e->vex_class == 1 || e->vex_class == 2);
+ assert(e->vex_class == 1 || e->vex_class == 2 || e->vex_class == 4);
if (s->prefix & (PREFIX_REPZ | PREFIX_REPNZ)) {
- e->vex_class = 3;
+ e->vex_class = e->vex_class < 4 ? 3 : 5;
if (s->vex_l) {
goto illegal;
}
--
2.38.1
reply other threads:[~2023-01-09 16:17 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230109161635.276783-1-pbonzini@redhat.com \
--to=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=richard.henderson@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).